PROJECT SUMMARY/ABSTRACT Coronary artery disease (CAD) remains the main cause of morbidity and mortality in the United States. Cardiac CT provides fast non-invasive assessment of CAD with a high sensitivity and negative predictive value ? provided that the lumen can be visualized. However, heavily calcified or stented coronary segments are non- assessable, precluding non-invasive diagnosis of flow-limiting coronary plaques in an estimated 2 million U.S. adults. In addition, the spatial resolution of state-of-the-art CT systems is insufficient for robust visualization of features associated with high risk plaques. Further, while CT can quantitatively evaluate the impact of obstructive CAD on myocardial function using dynamic perfusion imaging, this requires relatively high patient radiation doses, which has limited widespread adoption. Considering the high personal and societal cost of CAD, robust, accurate, non-invasive imaging of calcified and stented coronary arteries, high-risk plaque features, and myocardial perfusion defects in a single, low-radiation-dose exam is critically needed. Built by Siemens Healthcare, a first-of-its-kind, whole-body, photon-counting-detector (PCD) CT system was installed in 2014 at the Mayo Clinic. With support from NIH award EB016966, we showed that the increased iodine contrast-to-noise ratio, decreased electronic noise, spectral imaging capabilities, and improved spatial resolution of PCD-CT relative to commercial CT enabled us to accurately measure increased vasa vasorum density in injured swine carotid arterial walls, demonstrating the exceptional potential of PCD-CT in vascular imaging. Because this system lacks cardiac imaging capabilities, our objective is to develop and validate a PCD dual-source (DS) CT system and novel imaging algorithms to accurately assess CAD in humans, especially in patients with heavily calcified, stented, or high-risk plaques, and to identify patients with myocardial perfusion defects. Our premise is that the established benefits of PCD-CT, used with a DS geometry and advanced noise reduction and material decomposition algorithms, can meet these objectives. Our proposal is significant in many ways: the technology developments will benefit all of CT imaging; robust, accurate, non-invasive imaging of calcified and stented coronary arteries, high-risk plaque features, and myocardial perfusion defects in a single, low-radiation-dose exam will obviate the need for additional imaging, reducing the overall time and cost to comprehensively evaluate CAD and its clinical significance. To extend the demonstrated benefits of PCDs to cardiac CT will require numerous physics, engineering, and algorithm innovations, including novel noise reduction and material decomposition algorithms using energy, spatial and temporal domain redundancies, as well as deep learning. These advances will culminate in a large clinical study to demonstrate not merely that the images are ?better,? as is so often done, but that PCD-DSCT provides clinically-significant improvements in the diagnosis and management of patients with suspected CAD.